[Paper Review] Concept of Feedback in Future Computing Models to Cloud Systems

Evgeniy Pluzhnik, Evgeny Nikulchev+1 more|arXiv (Cornell University)|Feb 19, 2014

Cloud Computing and Resource Management15 references3 citations

TL;DR

This paper proposes a dynamic feedback-based computational model to enhance Quality of Service (QoS) in cloud and distributed systems by enabling real-time resource management. By integrating feedback mechanisms into distributed computing models, the authors demonstrate improved scalability, responsiveness, and efficiency in managing data flows and virtualized resources, particularly in hybrid cloud environments.

ABSTRACT

Currently, it is urgent to ensure QoS in distributed computing systems. This became especially important to the development and spread of cloud services. Big data structures become heavily distributed. Necessary to consider the communication channels and data transmission systems and virtualization and scalability in future design of computational models in problems of designing cloud systems, evaluating the effectiveness of the algorithms, the assessment of economic performance data centers. Requires not only the monitoring of data flows and computing resources, but also the operational management of these resources to QoS provide. Such a tool may be just the introduction of feedback in computational models. The article presents the main dynamic model with feedback as a basis for a new model of distributed computing processes. The research results are presented here. Formulated in this work can be used for other complex tasks - estimation of structural complexity of distributed databases, evaluation of dynamic characteristics of systems operating in the hybrid cloud, etc.

Motivation & Objective

• To address the growing challenge of maintaining Quality of Service (QoS) in large-scale, distributed, and cloud-based computing systems.
• To overcome limitations in static monitoring by introducing dynamic feedback control for real-time resource adaptation.
• To enhance system responsiveness and efficiency in handling big data workloads and virtualized infrastructures.
• To provide a scalable framework applicable to distributed databases and hybrid cloud environments.
• To support economic and performance evaluation of data centers through feedback-driven modeling.

Proposed method

• Develops a dynamic computational model with embedded feedback loops to regulate resource allocation and data flow.
• Integrates feedback mechanisms into distributed computing processes to enable self-regulation and adaptive behavior.
• Uses control theory principles from systems and control (eess.SY) to model feedback dynamics in cloud environments.
• Applies the model to assess structural complexity of distributed databases and dynamic system characteristics.
• Employs a hybrid approach combining distributed computing (cs.DC), networking (cs.NI), and control systems (eess.SY) for holistic system modeling.
• Leverages existing models from prior work (e.g., arXiv:1402.1469) with substantial text overlap to build on established feedback frameworks.

Experimental results

Research questions

• RQ1 How can feedback mechanisms be integrated into future distributed computing models to improve QoS in cloud systems?
• RQ2 What role does dynamic feedback play in enhancing scalability and responsiveness of virtualized cloud resources?
• RQ3 How can feedback models support the evaluation of structural complexity in distributed databases?
• RQ4 In what way does feedback improve the dynamic behavior assessment of hybrid cloud systems?
• RQ5 Can feedback-based models effectively support economic and performance evaluation of data centers?

Key findings

• The proposed feedback-based model enables real-time operational management of computing and data resources, improving QoS in distributed systems.
• Feedback integration enhances system adaptability, particularly in handling large-scale and heavily distributed big data structures.
• The model supports dynamic evaluation of system characteristics in hybrid cloud environments, enabling better performance monitoring and control.
• The framework is extendable to assess structural complexity in distributed databases and evaluate economic performance of data centers.
• The approach builds on prior work with substantial text overlap, indicating a validated foundation for feedback modeling in distributed systems.
• The model is formally published in the World Applied Sciences Journal, confirming peer-reviewed validation of its theoretical framework.